MU-MIMO Scheduling via Approximated Metric Sorting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multi-user MIMO systems face high complexity in optimizing user pairing and precoding, leading to sub-optimal performance due to static user pairing and inadequate consideration of channel geometry, resulting in reduced spectral efficiency and increased computational burden.
Innovation Solution
A method that calculates an approximated scheduling metric for each combination of users and spatial streams, sorting them to prioritize combinations with higher real user capacities, reducing computational complexity by avoiding unnecessary local scheduling metric calculations and focusing on interference-aware precoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If close-to-optimal MIMO precoding determination is performed using matrix manipulations and optimization steps, then spectral efficiency and signal quality are improved, but computational complexity increases significantly
Solution Approach 1:
The precoding process is segmented into two distinct stages: a low-complexity user pairing stage that uses simplified metrics to identify promising user combinations, and a high-complexity precoding stage that is only executed for the selected pairs. This segmentation allows the system to achieve good signal quality through the second stage while keeping overall computational complexity low by filtering out most combinations in the first stage.
Solution Approach 2:
The patent applies preliminary action by performing user pairing and preliminary metric calculation before the actual precoding operation. The scheduler calculates approximated scheduling metrics for all possible user pairs in advance, sorts them, and only performs the complex precoding determination on the top candidates. This preliminary sorting action significantly reduces the number of full optimization steps required.
2Device complexity
If static user pairing based on long term approach is used, then computational complexity is reduced, but spectral efficiency decreases due to missed opportunities to create adequate MU-MIMO channels
Solution Approach 1:
The patent introduces dynamics by making the user pairing strategy adaptive rather than static. The scheduler dynamically selects user pairs based on current channel conditions and real-time scheduling metrics, allowing the system to exploit temporal variations in channel geometry. This dynamic approach enables the system to capture transient opportunities for high spectral efficiency while maintaining manageable complexity through the two-stage process.
Solution Approach 2:
The patent changes the parameters used for user pairing from static long-term averages to dynamic instantaneous metrics that reflect current channel conditions. By using approximated scheduling metrics that incorporate current channel state information and scheduling objectives, the system adapts to changing conditions and captures temporal variations in channel geometry, thereby improving spectral efficiency without excessive complexity.
3Device complexity
If user selection is based on maximal interference free capacity, then low complexity is achieved, but the choice of first user can drastically reduce second user capacity and overall performance
Solution Approach 1:
The patent applies preliminary action by calculating and sorting approximated scheduling metrics for all possible user pairs before making selection decisions. This preliminary sorting ensures that the system evaluates multiple potential pairs and their interactions before final selection, preventing the greedy mistake of selecting users based solely on individual capacity without considering pairwise interactions.
Solution Approach 2:
The patent uses partial action by evaluating only the top-k user pairs based on approximated metrics rather than exhaustively checking all possibilities. This partial evaluation approach maintains low computational complexity while sufficiently capturing the trade-offs between users. The system performs enough analysis to identify the most promising pairs without the excessive computation of full exhaustive search.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The present invention relates to a method of scheduling. in the context of Multiuser MIMO techniques, in a cellular system comprising a plurality of antenna means transmitting a plurality of spatial streams for serving a plurality of users having respective capacities. The method comprises: - calculating an approximated scheduling metric for each possible combination of users among said plurality of users, and associated spatial streams among said plurality of spatial streams, the approximated scheduling metric being a function of user capacities without taking into account interference; - sorting the approximated scheduling metrics in descending order so as to obtain a classification of said possible combinations and calculating a local scheduling metric for the first combination in said classification, the local scheduling metric being a function of real user capacities taking into account interference, and calculated from the MIMO precoder; - then, for each following combination in the classification, comparing the approximated metric of the following combination to a preceding calculated local scheduling metric, and if the approximated metric of the following combination is higher than the preceding calculated local scheduling metric, then the local scheduling metric of said following combination is calculated and then replaces said preceding calculated local scheduling metric ; if the approximated metric of the following combination is lower than or equal to said preceding calculated local scheduling metric, defining the combination providing the highest calculated local scheduling metric as the best combination of users and associated spatial streams. The method is carried out for each of the radio resources on which a MU-MIMO transmission or reception may take place.